In our Center research evaluating contingency management (CM) that provides opportunities to win prizes with an average expense of under $250 per patient, we have noted that that the efficacy of CM may be dependent on the status of the individual as they initiate treatment. Patients who begin treatment with a cocaine-positive urinalysis result tend to drop out of treatment prematurely and to continue using while in treatment. CM interventions have been efficacious in reducing drug use in this subgroup, and the effects were magnitude dependent. However, we have thus far only tested up to a maximum of $250 in prizes, and larger magnitudes may further improve outcomes. One purpose of this proposal is to examine the efficacy of an enhanced CM procedure, in which increased frequencies of prize winnings are provided during initial periods of abstinence. Cocaine-dependent patients beginning treatment with a cocaine-positive urine sample (N=120) will be randomly assigned to one of three conditions: (a) standard, non-CM treatment, (b) standard treatment plus CM with an expected probability of winning about $250 in prizes, or (c) standard treatment plus CM with an expected probability of winning about $560 in prizes. We have also found that patients who present to treatment with cocaine-negative samples generally remit negative samples throughout their time in treatment, regardless of whether they received a non-CM or a CM treatment contingent upon abstinence. Thus, we will also conduct a parallel study that will assess whether simply reinforcing attendance at treatment enhances retention and improves long-term outcomes in this subgroup. Cocaine-dependent patients (N=330) initiating treatment with a cocaine-negative urine sample will be randomly assigned to one of three conditions: (a) standard treatment without CM or (b) standard treatment plus CM with an expected probability of winning about $250 worth of prizes contingent upon cocaine abstinence, or (d) standard treatment plus CM with an expected probability of winning about $250 worth of prizes contingent upon treatment attendance. Together, these studies will address the conditions under which lower and higher cost prize CM procedures may improve outcomes of cocaine-dependent patients.